[VPlan] Handle scalarized values in VPTransformState. (3201274d) · Commits · llvm-doe / llvm-project

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

+9 −3

Original line number	Diff line number	Diff line
		@@ -7718,9 +7718,15 @@ void LoopVectorizationPlanner::executePlan(InnerLoopVectorizer &ILV,

		assert(BestVF.hasValue() && "Vectorization Factor is missing");

		VPTransformState State{*BestVF, BestUF, LI,
		DT, ILV.Builder, ILV.VectorLoopValueMap,
		&ILV, CallbackILV};
		VPTransformState State{*BestVF,
		BestUF,
		OrigLoop,
		LI,
		DT,
		ILV.Builder,
		ILV.VectorLoopValueMap,
		&ILV,
		CallbackILV};
		State.CFG.PrevBB = ILV.createVectorizedLoopSkeleton();
		State.TripCount = ILV.getOrCreateTripCount(nullptr);
		State.CanonicalIV = ILV.Induction;

llvm/lib/Transforms/Vectorize/VPlan.cpp

+21 −0

Original line number	Diff line number	Diff line
		@@ -216,6 +216,27 @@ VPBasicBlock::iterator VPBasicBlock::getFirstNonPhi() {
		return It;
		}

		Value VPTransformState::get(VPValue Def, const VPIteration &Instance) {
		if (!Def->getDef() && OrigLoop->isLoopInvariant(Def->getLiveInIRValue()))
		return Def->getLiveInIRValue();

		if (hasScalarValue(Def, Instance))
		return Data.PerPartScalars[Def][Instance.Part][Instance.Lane];

		if (hasVectorValue(Def, Instance.Part)) {
		assert(Data.PerPartOutput.count(Def));
		auto *VecPart = Data.PerPartOutput[Def][Instance.Part];
		if (!VecPart->getType()->isVectorTy()) {
		assert(Instance.Lane == 0 && "cannot get lane > 0 for scalar");
		return VecPart;
		}
		// TODO: Cache created scalar values.
		return Builder.CreateExtractElement(VecPart,
		Builder.getInt32(Instance.Lane));
		}
		return Callback.getOrCreateScalarValue(VPValue2Value[Def], Instance);
		}

		BasicBlock *
		VPBasicBlock::createEmptyBasicBlock(VPTransformState::CFGState &CFG) {
		// BB stands for IR BasicBlocks. VPBB stands for VPlan VPBasicBlocks.

llvm/lib/Transforms/Vectorize/VPlan.h

+34 −20

Original line number	Diff line number	Diff line
		@@ -246,12 +246,12 @@ struct VPCallback {
		/// VPTransformState holds information passed down when "executing" a VPlan,
		/// needed for generating the output IR.
		struct VPTransformState {
		VPTransformState(ElementCount VF, unsigned UF, LoopInfo *LI,
		VPTransformState(ElementCount VF, unsigned UF, Loop OrigLoop, LoopInfo LI,
		DominatorTree *DT, IRBuilder<> &Builder,
		VectorizerValueMap &ValueMap, InnerLoopVectorizer *ILV,
		VPCallback &Callback)
		: VF(VF), UF(UF), Instance(), LI(LI), DT(DT), Builder(Builder),
		ValueMap(ValueMap), ILV(ILV), Callback(Callback) {}
		: VF(VF), UF(UF), Instance(), OrigLoop(OrigLoop), LI(LI), DT(DT),
		Builder(Builder), ValueMap(ValueMap), ILV(ILV), Callback(Callback) {}

		/// The chosen Vectorization and Unroll Factors of the loop being vectorized.
		ElementCount VF;
		@@ -269,6 +269,9 @@ struct VPTransformState {
		typedef SmallVector<Value *, 2> PerPartValuesTy;

		DenseMap<VPValue *, PerPartValuesTy> PerPartOutput;

		using ScalarsPerPartValuesTy = SmallVector<SmallVector<Value *, 4>, 2>;
		DenseMap<VPValue *, ScalarsPerPartValuesTy> PerPartScalars;
		} Data;

		/// Get the generated Value for a given VPValue and a given Part. Note that
		@@ -285,24 +288,21 @@ struct VPTransformState {
		}

		/// Get the generated Value for a given VPValue and given Part and Lane.
		Value get(VPValue Def, const VPIteration &Instance) {
		// If the Def is managed directly by VPTransformState, extract the lane from
		// the relevant part. Note that currently only VPInstructions and external
		// defs are managed by VPTransformState. Other Defs are still created by ILV
		// and managed in its ValueMap. For those this method currently just
		// delegates the call to ILV below.
		if (Data.PerPartOutput.count(Def)) {
		auto *VecPart = Data.PerPartOutput[Def][Instance.Part];
		if (!VecPart->getType()->isVectorTy()) {
		assert(Instance.Lane == 0 && "cannot get lane > 0 for scalar");
		return VecPart;
		}
		// TODO: Cache created scalar values.
		return Builder.CreateExtractElement(VecPart,
		Builder.getInt32(Instance.Lane));
		Value get(VPValue Def, const VPIteration &Instance);

		bool hasVectorValue(VPValue *Def, unsigned Part) {
		auto I = Data.PerPartOutput.find(Def);
		return I != Data.PerPartOutput.end() && Part < I->second.size() &&
		I->second[Part];
		}

		return Callback.getOrCreateScalarValue(VPValue2Value[Def], Instance);
		bool hasScalarValue(VPValue *Def, VPIteration Instance) {
		auto I = Data.PerPartScalars.find(Def);
		if (I == Data.PerPartScalars.end())
		return false;
		return Instance.Part < I->second.size() &&
		Instance.Lane < I->second[Instance.Part].size() &&
		I->second[Instance.Part][Instance.Lane];
		}

		/// Set the generated Value for a given VPValue and a given Part.
		@@ -315,6 +315,17 @@ struct VPTransformState {
		}
		void set(VPValue Def, Value IRDef, Value *V, unsigned Part);

		void set(VPValue Def, Value V, const VPIteration &Instance) {
		auto Iter = Data.PerPartScalars.insert({Def, {}});
		auto &PerPartVec = Iter.first->second;
		while (PerPartVec.size() <= Instance.Part)
		PerPartVec.emplace_back();
		auto &Scalars = PerPartVec[Instance.Part];
		while (Scalars.size() <= Instance.Lane)
		Scalars.push_back(nullptr);
		Scalars[Instance.Lane] = V;
		}

		/// Hold state information used when constructing the CFG of the output IR,
		/// traversing the VPBasicBlocks and generating corresponding IR BasicBlocks.
		struct CFGState {
		@@ -340,6 +351,9 @@ struct VPTransformState {
		CFGState() = default;
		} CFG;

		/// Hold a pointer to the original loop.
		Loop *OrigLoop;

		/// Hold a pointer to LoopInfo to register new basic blocks in the loop.
		LoopInfo *LI;